The proposed studies are aimed at understanding how the HIV-1 Gag polyprotein and its constituent matrix (MA) and capsid (CA) domains function during viral replication. During the current funding period we (i) determined the structures of the N-terminal half and central portions of HIV-1 Gag, (ii) identified a structural switch in CA that appears to trigger capsid assembly, (iii) determined the structure of the myristoylated HIV-1 MA protein, (iv) identified a novel entropic myristyl switch mechanism that regulates membrane binding, (v) determined the structures of the intact capsid proteins from the related human T-cell leukemia (HTLV-I) and Rous Sarcoma (RSV) viruses, providing the first 3D structures of intact retroviral capsid proteins, and (vi) discovered the first antiviral inhibitors of HIV-1 capsid assembly. We are now poised to determine how Gag, MA and CA interact with each other and with viral and cellular constituents, and to improve the efficacy of antiviral agents that inhibit capsid assembly. First, to understand the mechanisms that regulate specific membrane binding and intracellular trafficking, interactions of MA with inositol phosphates (IPs) and the AP-3 trafficking complex will be studied. Preliminary studies indicate that IPs bind MA and trigger myristate exposure, which may explain how Gag is targeted to IP-rich membranes for assembly. MA mutations and phosphorylation events that affect nuclear localization will also be studied. Next, to gain insights into Gag-Gag interactions that promote virus assembly, we will determine the structures of chimeric Gag constructs containing trimer-promoting GCN4-derived peptides substituted for the N-terminal myristyl group. (Studies of myristoylated Gag trimers are precluded by higher-order aggregation at NMR concentrations). Preliminary studies indicate that GCN4-Gag constructs form well- behaved trimers with previously uncharacterized trimeric CA interfaces. Third, efforts will be made to improve the efficacy of capsid assembly inhibitors. We have now identified compounds with 500-fold improved binding affinities (Kd ~ 1 ^M) and >100-fold improved EC50 values (~20 (j,M) compared to our initial leads. Thermodynamic and structural information obtained for CA:inhibitor complexes will be used in conjunction with in vivo studies to identify more potent antiviral agents with potential clinical utility. Studies of multi-component systems are technically more challenging than our previous studies of isolated retroviral proteins and domains, but the potential rewards are significantly higher, and could lead to the development of new drugs for the treatment of emerging multi-drug resistant strains of HIV.